Understanding the Value of Transport Infrastructure ...

UNDERSTANDING THE VALUE OF TRANSPORT INFRASTRUCTURE
infrastructure as an input variable in an aggregate production function and measuring the
empirical relationship between transport infrastructure input and aggregate output allows
for a direct effect and an indirect effect (transport infrastructure boosting total factor
productivity and thereby output). Ideally, one would consider the relationship between
components of the transport network and growth, in order to assess how close the
existing network is to its growth-maximizing (or optimal) level.
It is important to note that the policy target for the policy maker is the optimality of the
network (i.e. a stock of assets). The annual investment (flow) into that network is just a
short-term decision variable meant to serve the broader policy target.
Transport infrastructure is part of the broader concept of public capital. The empirical
literature on the productivity of public capital has been summarized in Kamps (2005),
Jong-A-Pin and de Haan (2008), and Crafts (2009). The early empirical literature in the
late 1980s, using mostly US data, suggested that the growth impact of public capital was
very high. The econometric methodology of these early studies was subsequently
questioned, and the application of more appropriate methods led to lower estimates.
It is important to recognize that public capital is often used in these studies as a proxy
for infrastructure or, more specifically, transport infrastructure, given that a large share
of government investment (capital) comprises transport infrastructure, specifically roads.
However, as emphasized in Gonzalez-Alegre et al. (2008), these concepts are empirically
unequal, so conclusions based on data for overall government investment (capital) are
not immediately applicable for investment in transport infrastructure. The heterogeneity
of public capital across countries and time periods also explains why the results of
empirical analyses tend to vary significantly.
Box 4. Impact of road infrastructure on productivity of manufacturing in Finland
Uimonen and Tuovinen (2006) have estimated the impact of the road infrastructure on
costs and productivity of the manufacturing industry in Finland. The study uses new data
on the asset value (capital stock) of road infrastructure and applies an econometric
model to estimate the impact of the road network on manufacturing productivity.
According to the study, road infrastructure has had a significant impact on the
productivity of the manufacturing industry. A 1% increase in investment in road
infrastructure resulted in 0.3% savings in the short-term costs for manufacturing
industries. In addition, investments in infrastructure were clearly more productive than
investments in private capital.
As more time-series data have become available, it has also become possible to estimate
how the productivity of public capital may have changed over time. While this branch of
literature is still young, there is some evidence that the productivity of public capital has
been declining over time in advanced economies, notably in Europe. This tentative
conclusion is intuitively appealing insofar as one would indeed expect the growth impact
of, say, motorway construction to be large in the early phases of network development
as key hubs are connected. The more complete the network becomes, the lower the
average impact of yet another segment is likely to become. However, it is important to
point out again that even if the average impact may be low, individual projects may still
have a very high economic rate of return and may be worth pursuing even in the context
of well-developed networks.
© OECD/ITF 2013 31